Copy Link
Add to Bookmark
Report
Alife Digest Number 009
Artificial Life Digest, Number 9
Monday, March 26th 1990
Issue's Topics:
Emacs compatible
Initial investigations into Artificial Life (Long)
Genetic Algorithms and (non)-binary alphabets
SAB90 Call for Papers
----------------------------------------------------------------------
Date: Mon, 26 Mar 90 11:18:29 EST
From: Eric T. Freeman <efreeman>
Subject: Emacs compatible
This is our first attempt to create a emacs-compatible digest. Direct
all problems to alife-request@iuvax.cs.indiana.edu.
Thanks,
Eric Freeman
------------------------------
Date: 22 Mar 90 16:54:29 GMT (Thu)
From: Harold Thimbleby <hwt%compsci.stirling.ac.uk@nsfnet-relay.ac.uk>
Subject: Initial investigations into Artificial Life (Long)
I've posted this note a couple of times to alife, but I've not
seen it appear ... maybe it's got lost over the Atlantic.
I enclose the following jottings for comment, feedback, extension, etc.
I know the net has is having a fuss over spurious philosophical issues
--- I'd also be interested to hear what people FEEL about these sorts
of issues, whether they are relevant etc., as well as what they think
about their correctness!
Of course, helpful additions will be welcome!
Initial investigations into Artificial Life
Harold Thimbleby
Wednesday, February 21, 1990
1. If you watched a film of animals on television, you would
say that what you could actually see in front of you --- the
television --- was not alive in any sense, but you'd agree that
what was filmed was or had been alive. If now you watch a
computer screen showing something claimed to be AL, would you say
it was alive, or merely a film of something that may or may not
be alive?
2. Turing Completeness is neither necessary nor sufficient
for AL. The existence of (say) worms which, although they can
compute limited functions, cannot be organised into a general
computational device, shows that Turing Completeness is not
necessary for life. Insufficiency is easily shown by considering
an actual Turing Machine: whatever it was doing, no human
observer would want to say that it was alive! At best it would
use symbols and be too slow or too mysterious.
3. Real life dies. Does life embedded in silicon die? What
would the point of death be for AL?
4. The Turing Test does not help. Imagine the test between
a computer and a cat, with a human interrogator. The computer
merely has to do nothing, and then the human cannot tell the
computer from the cat. Therefore the computer is alive!? No,
the Turing Test is a refutation procedure: only when the
interrogator can see the difference do you know that the two
tested objects are distinguishable.
5. Consciousness is not sufficient for AL. A machine might
be able to pass the Turing Test, yet still be `just' a computer.
The Turing Test tests for intelligence, and presumably
consciousness, not for life.
6. We know too much about biological life, so we get
confused with such examples as viruses, plant seeds, eggs and so
on. What about a body undergoing autolytic decomposition: part
of it is alive! Perhaps we should define life as obviously
alive, and exclude things that are not observable to the naked
eye.
7. Computer viruses are certainly as alive as their
biological counterparts. An what is more, computer viruses are
out of control of their creators. They are far more alive than
any AL simulation that is stopped when its computer is switched
off.
8. What if a lump of rock on the beach was alive? How would
you know, and even if you did, there would be no consequences for
believing it. The rock might, of course, be silicon.
9. Life can be defined as a relationship with
uncontentiously alive beings, such as humans. Worms are alive
(by common consent). Many people who have never talked to worms
have a working relationship with computers. Therefore computers
are alive.
10. Life is a vague concept. The argument goes along the
well-worn Sorites paradox lines: A computing device with one
gate is clearly not alive. Adding a gate to a non-living
computing device does not make it alive. Therefore (by
induction) no number of gates is sufficient to make a computing
device alive. (The proof can be considerably strengthened by
making `one gate' any member of a set of appropriate components.)
11. Either you disagree with the last point, or you have to
agree that a single neuron (or any other biological building
brick) is alive. It follows that life is not an emergent
property, since it can be found in any identifiable biological
component.
12. A solution to the Sorites paradox is that the observer
permits several gates to be added all at once. ``It wasn't alive
then, but now it is.'' The corollary is that when putative life
is analysed at too low or too frequent a level of detail it never
appears to be alive.
13. Is AL accountable? If AL replicates autonomously, who is
responsible for the `wild' (or ferral) animals thereby created?
14. In British Law, you need a licence to propagate the
African Violet (a pleasant little flower), because you would be
competing against the originator of the African Violet strain you
worked from.
15. Destructiveness is easier to achieve than cooperation.
Is it a coincidence that the largest things at the Second
Artificial Life Workshop were destructive?
16. All advanced technology has had an unexpectedly large
impact on toys and recreation. (Contrast the first predictions
for the number of computers a country required with the number
that are now used in arcade games.) AL can likewise be expected
to have a major impact on toys. Well, at least AL should be fun.
17. If researchers can learn anything about life from AL,
then children ought to be able to learn how to be biologists,
ecologists, ethologists. Slogan: Make AL educational!
18. If AL turns out to be difficult, or even a non-starter,
at least we will have a greater appreciation just for being
alive.
19. Searle uses his Chinese Room thought experiment to argue
that computers could not understand Chinese (and hence by
implication other natural languages). He claims that he would
not be `conscious' of understanding Chinese. Here, I agree with
him. But there are two intriguing issues: first, Searle is
obviously the hardware implementing the software of his Chinese
rule book. Therefore, it is not surprising that he, as hardware,
is not conscious: the rule book is conscious, or at least some
combination of rule book plus Searle, rather than Searle alone,
is conscious. The second point is, granted Searle denies he
understands Chinese, but what happens when the outside-the-room
Chinese speaker asks the room whether `it' understands Chinese?
The only sensible answer, by assumption, is that the room replies
that, yes, it does!
20. My conclusion from Searle's experiment is that the
consciousness of part of an organism (in this case, a room, rule
book, and Searle himself) cannot be relied upon to form an
accurate image of the consciousness of the entire organism.
21. If Searle has a rule book, then I see no reason why a
computer could not implement its rules. First question: could an
outside observer tell that we have replaced Searle with a
computer (shades of the Turing Test)? The next question: whether
the computer is alive or not, or thinks itself to be alive or
not, is immaterial to the answers it gives to an outside Chinese
interrogator!
22. If the rule book allowed Searle to answer, ``What's your
favourite method of reproduction?'', the computer could answer
the same question --- with the same answer.
23. The interesting question for AL is whether the rule book
postulated by Searle can exist, even in principle.
24. Maybe be AL will evolve life more intelligent than we are
but there is no reason to suppose that intelligence and altruism
are in any way related. We may be much worse off, even though
there is 'life' that could, but only in principle, help us solve
our tricky survival problems.
To be continued
------------------------------
Date: Fri, 23 Mar 90 17:59:54 +0100
From: christer@cs.umu.se
Subject: Genetic Algorithms and (non)-binary alphabets
Is it really that important to use a binary alphabet to represent individuals
in genetic algorithms as many prominent GA people claim it to be?
Schaffer writes (in Davis' "Genetic Algorithms and Simulated Annealing" p.90):
"There is nothing sacred about a binary character set. In principle any
character set may be used. However, there are reasons for preferring binary.
The arguments have been given by Holland (Holland 1975, p.71) and by Smith
(Smith 1980, p.56) and some empirical evidence for the correctness of these
arguments has been presented by Schaffer (Schaffer 1984, p.107)."
I don't have access to any of the works mentioned (Hollands book is impossible
to come by - anyone has a copy he/she wants to sell?). Could someone please
summarize these reasons?
While David Goldberg (in his book "Genetic Algorithms in Search, Optimization
and Machine Learning") seems to take every opportunity to condemn the use of
non-binary alphabets. He also states 'the principle of minimal alphabets' as:
"The user should select the smallest alphabet that permits a natural expression
of the problem."
What if I'm encoding a programming language (eg LISP) in binary, I still
have to combine bits in order to decode an instruction - why not represent
the instructions and their arguments as integers right away? Especially if I
only have 5 instructions, then I would need to add 3 NOP instructions (or
whatever) to ensure that I have all 8 instructions that the 3 bits (needed
to encode the 5 instructions) decode for.
When is it reasonable to use a binary alphabet and when isn't it?
/Christer
| Christer Ericson Internet: christer@cs.umu.se [130.239.1.101] |
| Department of Computer Science, University of Umea, S-90187 UMEA, Sweden |
| "Track 0 sector 0 must *always* load into page 8!" -Krakowicz' first law |
------------------------------
Subject: SAB90 Call for Papers
Date: Mon, 26 Mar 90 09:42:30 EST
From: Stewart Wilson <wilson@think.com>
March 26, 1990
Dear Colleague,
Enclosed is the Call for Papers for the conference "Simulation of
Adaptive Behavior: From Animals to Animats" that will be held in Paris
in September, 1990, and is chaired by Jean-Arcady Meyer and me. We believe
that study and simulation of adaptive animal behavior is an important and
potentially fruitful new trend in artificial intelligence. The idea of the
conference is to bring together people who are trying to understand adaptive
behavior in real animals with people attempting to create adaptive artificial
animals. Each group has much to learn from the other. Since each group also
does considerable modeling, we have emphasized computer simulations-or,
where possible, physical realizations-as a good vehicle for presenting
and comparing ideas.
. From initial reactions, the conference has an exciting and timely
theme, and we have been fortunate to assemble an outstanding program
committee. We very much hope you will contribute a paper and will be able
to attend. Papers will be reviewed by members of the program committee
and will be published in a proceedings volume available shortly after
the conference. For information, when it is ready, on arrangements for
attending the conference, kindly confirm this message and include your
postal address. We look forward to seeing you "a Paris"!
Yours sincerely,
Stewart Wilson
Call for Papers
SIMULATION OF ADAPTIVE BEHAVIOR: FROM ANIMALS TO ANIMATS
An International Conference to be held in Paris
September 24-28, 1990
The object of the conference is to bring together researchers in
ethology, ecology, cybernetics, artificial intelligence, robotics,
and related fields so as to further our understanding of the behaviors
and underlying mechanisms that allow animals and, potentially,
robots to adapt and survive in uncertain environments.
The conference will focus particularly on simulation models in
order to help characterize and compare various organizational principles
or architectures capable of inducing adaptive behavior in real or
artificial animals.
Contact among scientists from diverse disciplines should contribute
to better appreciation of each other's approaches and vocabularies,
to cross-fertilization of fundamental and applied research, and
to defining objectives, constraints, and challenges for future work.
Contributions treating any of the following topics from the
perspective of adaptive behavior will receive special emphasis.
Individual and collective behaviors Autonomous robots
Action selection and behavioral Hierarchical and parallel organizations
sequences Self organization of behavioral
Conditioning, learning and induction modules
Neural correlates of behavior Problem solving and planning
Perception and motor control Goal directed behavior
Motivation and emotion Neural networks and classifier
Behavioral ontogeny systems
Cognitive maps and internal Emergent structures and behaviors
world models
Authors are requested to send two copies (hard copy only) of a
full paper to each of the Conference chairmen:
Jean-Arcady MEYER Stewart WILSON
Groupe de Bioinformatique The Rowland Institute for Science
URA686.Ecole Normale Superieure 100 Cambridge Parkway
46 rue d'Ulm Cambridge, MA 02142
75230 Paris Cedex 05 USA
France
e-mail: meyer@FRULM63.bitnet e-mail: wilson@think.com
meyer@hermes.ens.fr
A brief preliminary letter to one chairman indicating the intention to
participate--with the tentative title of the intended paper and a list
of the topics addressed--would be appreciated for planning purposes.
For conference information, please also contact one of the chairmen.
Conference committee:
Conference Chair J.A. Meyer, S. Wilson
Organizing Committee Groupe de BioInformatique.ENS.France.
and local arrangements A. Guillot, J.A. Meyer, P. Tarroux,
P. Vincens
Program Committee L. Booker, USA R. Brooks, USA
P. Colgan, Canada P. Greussay, France
D. McFarland, UK L. Steels, Belgium
R. Sutton, USA F. Toates, UK
D. Waltz, USA
Official Language: English
Important Dates
31 May 90 Submissions must be received by the chairmen
30 June 90 Notification of acceptance or rejection
31 August 90 Camera ready revised versions due
24-28 September 90 Conference dates
End of ALife Digest
********************************
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=---=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
= Artificial Life Distribution List =
= =
= All submissions for distribution to: alife@iuvax.cs.indiana.edu =
= All list subscriber additions, deletions, or administrative details to: =
= alife-request@iuvax.cs.indiana.edu =
= All software, tech reports to Alife depository through =
= anonymous ftp at iuvax.cs.indiana.edu in ~ftp/pub/alife =
= =
= List maintainers: Elisabeth Freeman, Eric Freeman, Marek Lugowski =
= Artificial Life Research Group, Indiana University =
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=---=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
End of Alife Digest
********************************
